Onboarding

A practical checklist for getting set up to develop on cnbs-predictor.

1. Environment

The project uses a conda environment named nbs_env.

git clone https://github.com/great-lakes-ai-lab/cnbs-predictor.git
cd cnbs-predictor
conda env create -f requirements/environment.yml
conda activate nbs_env
python -m ipykernel install --user --name nbs_env --display-name "Python (nbs_env)"

requirements/environment.yml is the full environment (includes the TensorFlow stack used for training). There is also a lighter requirements/environment-test.yml — no TensorFlow — used by the automated test suite and continuous integration.

2. Running the tests

# Fast, offline tests (what CI runs on every PR):
pytest -m "not network"

# The live NOAA/AWS reachability checks (run on a schedule in CI):
pytest -m network

Tests that hit live external endpoints are marked @pytest.mark.network so they can be excluded from the default run — a NOAA outage shouldn't block a PR. See CONTRIBUTING.md for the full rationale.

3. Branch naming

From CONTRIBUTING.md, branches use a type prefix + initials + issue number + short description:

• feature/ — e.g. feature/dj_123_add-forecasting-model
• bugfix/ — e.g. bugfix/dj_456_fix-csv-import
• docs/ — e.g. docs/dj_101_update-readme
• test/ — e.g. test/dj_102_add-unit-tests

4. Continuous integration

On every pull request to main, GitHub Actions builds the lightweight test environment and runs the offline suite. The check grows automatically as new test files land — no workflow changes needed. The live-network tests run on a daily schedule instead.

5. Documentation

• Reference docs (install/usage/API) are built with Sphinx from docs/sphinx/ and published on Read the Docs. To build locally:

  pip install -r docs/sphinx/requirements.txt
  cd docs/sphinx && make html

• This Wiki holds onboarding, design notes, FAQ, and the decision log.